Tion 5. Mapping from Planning Frame to Articulator Frame Articulator Position Vector Auditory Processing Planning Position Vector a Modeling Framework for Speech Motor Development and Kinematic Articulator Control
نویسندگان
چکیده
that the same speaker will often use widely different articulator configurations to produce /r/ in different contexts. Ongoing research is addressing the addition of F3 to the planning space and the addition of consonants to the model's phonemic repertoire. It is believed that consonants, unlike vowels and /r/, may require the incorporation of constriction information in the target specification. Therefore, a hybrid planning space including both formants and constrictions will be investigated. A neural network model of speech acquisition and motor equivalent speech production. " Biological Cybernetics, vol. 72, pp. 43-53. [2] Guenther, F.H. (in press), " Speech sound acquisition, coarticulation, and rate effects in a neural network model of speech production. " Psychological Review. [3] Maeda, S. (1990), " Compensatory articulation during speech: Evidence from the analysis and synthesis of vocal-tract shapes using an articulatory model. " In Hardcastle W. The role of contrast in limiting vowel-to-vowel coar-ticulation in different languages. " Journal of the Acoustical Society of America, (1993), " The interplay between prosodic structure and coarticu-lation. " (1979), " Formant frequencies of some fixed-mandible vowels and a model of speech motor programming by predictive simulation. " " Trading relations between tongue-body raising and lip rounding in production of the vowel /u/: A pilot 'motor equiva-lence' study. A dialect study of American r's by x-ray motion picture. A self-organizing neu-ral model of motor equivalent reaching and tool use by a multijoint arm. " by the weights between the Speech Sound Map and the Planning Direction Vector stages of the model. The Planning Direction Vector activity then represents the desired movement direction in for-mant space. These desired formant changes are transformed into articulator movements through the direction-to-direction mapping manifested by the weights projecting from the Planning Direction Vector to the Articulator Direction Vector. These movements result in the formant trajectories shown in the top half of Figure 3. The solid arrow is the trajectory produced when /d/ precedes /r/, and the dashed arrow is the trajectory produced when /g/ precedes /r/. The important thing to note is that the direction-to-direction mapping transforms these formant trajectories, which go to a single target region in formant space, into articulator trajectories that end up at different sub-regions in articu-lator space. The articulator space trajec-tories are indicated by the arrows in the bottom half of Figure 3. (Because this plot represents just two of the seven articulator dimensions, the trajectories are only approximate.) …
منابع مشابه
A Modeling Framework for Speech Motor Development and Kinematic Articulator Control
This paper presents three hypotheses that are central to a computational model of speech production: (1) Sound targets take the form of regions, rather than points, in a planning reference frame. (2) The planning frame is more acoustic-like than the frames used in most recent models. (3) A direction-to-direction mapping transforms planned trajectories into articulator movements. These hypothese...
متن کاملA Neural Model of Speech Production and Its Application to Studies of the Role of Auditory Feedback in Speech
This paper describes a neural model of speech production and perception-production interactions. This model has been developed to account for a wide variety of experimental data, ranging from kinematic analyses of articulator movements to functional imaging studies of the human brain. Hypothesized neural correlates of the model’s components have been identified to facilitate testing of model pr...
متن کاملAn Unsupervised Method for Learning to Track Tongue Position from an Acoustic Signal*
A procedure is demonstrated for learning to recover the relative positions of simulated articulators from speech signals generated by articulatory synthesis. The algorithm learns without supervision, that is, it does not require infonnation about which articulator configurations created the acoustic infonnation in the training set. The procedure consists of vector quantizing short time windows ...
متن کاملA theoretical investigation of reference frames for the planning of speech movements.
Does the speech motor control system use invariant vocal tract shape targets when producing vowels and semivowels? A 4-part theoretical treatment favoring models whose only invariant targets are regions in auditory perceptual space over models that posit invariant constriction targets is presented. Auditory target regions are hypothesized to arise during development as an emergent property of n...
متن کاملAttention fine-tunes auditory-motor processing of speech sounds.
The earliest stages of cortical processing of speech sounds take place in the auditory cortex. Transcranial magnetic stimulation (TMS) studies have provided evidence that the human articulatory motor cortex contributes also to speech processing. For example, stimulation of the motor lip representation influences specifically discrimination of lip-articulated speech sounds. However, the timing o...
متن کامل